Possible to avoid first Buffer " *net-fsm*" has a running process?

[dotmilk]

This occurs every time I start up everything and send the ping. Often I don't notice the message and prompt. By the time I do and press Y, I get:

error in process filter: epmd-recv-port-resp: Assertion failed: (> (length arg) 2)
error in process filter: Assertion failed: (> (length arg) 2)

[dotmilk]

And as a related question: why is the fsm being killed on ping to begin with?

[bdionne]

Yea this is a bit annoying, so I dug into it a bit. If I comment out this line in net-fsm the problem goes away.

What seems to be happening is erl-connect makes a call to fsm-connect to connect to epmd, passing a success continuation that makes another call to fsm-connect to connect to the node specified. The first FSM terminates and kills the buffer, giving that message.

Commenting out that line works for me but I don't know enough about this code yet to submit a PR or grok what I might have broken. Looking at fsm-connect, fsm-open-socket, and fsm-attach, it appears to want to reuse the same process and switch the buffers. I'll poke at it some more.

[massemanet]

this is fixed in http://www.github.com/massemanet/distel/2018 branch.